Molding CNNs for text: Non-linear, non-consecutive convolutions
Author(s)Lei, Tao; Barzilay, Regina; Jaakkola, Tommi S.
MetadataShow full item record
The success of deep learning often derives from well-chosen operational building blocks. In this work, we revise the temporal convolution operation in CNNs to better adapt it to text processing. Instead of concatenating word representations, we appeal to tensor algebra and use low-rank n-gram tensors to directly exploit interactions between words already at the convolution stage. Moreover, we extend the n-gram convolution to non-consecutive words to recognize patterns with intervening words. Through a combination of low-rank tensors, and pattern weighting, we can efficiently evaluate the resulting convolution operation via dynamic programming. We test the resulting architecture on standard sentiment classification and news categorization tasks.Our model achieves state-of-the-art performance both in terms of accuracy and training speed. For instance, we obtain 51.2% accuracy on the fine-grained sentiment classification task.
DepartmentMassachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing
Association for Computational Linguistics
Lei, Tao, Regina Barzilay, and Tommi Jaakkola. "Molding CNNs for text: Non-linear, non-consecutive convolutions." Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 1565–1575, Lisbon, Portugal, 17-21 September 2015., pp.1565-1575.